The investigation of HLA microsatellites influence in predisposition to sarcoidosis among Croatians

Transcription

1 Original article: laboratory research SARCOIDOSIS VASCULITIS AND DIFFUSE LUNG DISEASES 2011; 28; Mattioli 1885 The investigation of HLA microsatellites influence in predisposition to sarcoidosis among Croatians Z. Grubic' 1, T. Peroš-Golubičic' 2, K. Š tingl 1, R. Ž unec 1 1 Tissue Typing Centre, University Hospital Centre Zagreb; 2 University Hospital for Lung Disease Jordanovac, Zagreb, Croatia Abstract. The aim of the present study was to analyze the distribution of HLA alleles (A, B, DRB1, DQB1) and HLA microsatellite alleles (TNFa, TNFb, TNFd, D6S273, D6S1014) in the Croatian patients with acute (N=93), as well as chronic sarcoidosis (N=40), in comparison to healthy controls (N=177), and investigate whether the polymorphism within the HLA region could be associated with different forms of sarcoidosis. Genomic DNA was isolated from peripheral blood. Patients were analyzed for HLA class I loci (A, B) by serology, while PCR-SSP method was used for HLA class II loci (DRB1, DQB1). Five HLA microsatellites were analyzed by PCR and electrophoresis in an automated sequencer. No significant deviation in the distribution of frequencies at HLA class I alleles was observed between the two patients subgroups and controls. Regarding the HLA class II alleles, a statistically significant increase in frequency of HLA-DRB1*03 and DQB1*0201 allele was found among patients with acute sarcoidosis in comparison to controls as well as in comparison to patients with chronic sarcoidosis. The same finding was observed for HLA-DRB1*03/DQB1*0201 haplotype (Pcorr=0.0168; OR=2.83). In the group of patients with chronic sarcoidosis DRB1*11 (P=0.0219; OR=2.44), DRB1*15 (P=0.0414; OR=2.47) demonstrated statistically significant difference in comparison to controls only, while a lower frequency of DRB1*13 (P=0.0156; OR=0.24) in this group was statistically significant when compared to both patients with acute sarcoidosis and controls. None of the alleles at TNFa microsatellite showed significant difference in distribution among both subgroups of patients and controls. Significant difference between patients with acute form of disease and controls was found for the following alleles: TNFd-2 (Pcorr= ; OR=4.89), D6S273-7 (Pcorr=0.0213; OR=2.96), and D6S (Pcorr=0.0028; OR=3.97). On the other hand, patients with chronic sarcoidosis differed from control subjects for D6S (Pcorr=0.0296; OR=8.35) allele. This study suggests the existence of an association of non-hla markers with sarcoidosis and the involvement of the region between HLA-DQB1 and D6S273 loci in its pathophysiology. (Sarcoidosis Vasc Diffuse Lung Dis 2011; 28: 18-26) Key words: sarcoidosis, HLA, TNF, D6S273, D6S1014, Croatians Received: 29 January 2010 Accepted after Revision: 22 November 2010 Correspondence: Zorana Grubić Tissue Typing Centre University Hospital Centre Zagreb Kispaticeva 12, Zagreb Croatia Tel Fax: zgrubic@kbc-zagreb.hr Introduction Sarcoidosis is a multisystem granulomatous disease of unknown etiology characterized by an accumulation of activated, proliferating T lymphocytes and mononuclear phagocytes that predominantly affects the lungs (1). Disease occurs among men and women of all races and ages. A few hypotheses attempted to explain the cause of sarcoidosis, but the

2 The investigation of HLA microsatellites influence in predisposition to sarcoidosis among Croatians 19 most widely accepted theory is that a T-helper cell/macrophage alveolitis in pulmonary sarcoidosis is caused by an unknown stimulus which induces the alveolar macrophages and lymphocytes to release mediators such as tumor necrosis factor-α (TNF-α) and interleukin-2 (IL-2) (2). Most studies on the association between sarcoidosis and genetic factors have focused on the HLA genes (3-6). The strongest association has been found with HLA-A and DRB1 alleles (7-9). We have also reported about the association between HLA-B8, -DRB1*0301 and -DQB1*0201 alleles and different onset of sarcoidosis in the Croatian population (7). A role of this haplotype has been implicated in many autoimmune diseases. However, approximately 40% patients with the acute onset of the disease have a predisposing HLA-DRB1*0301/DQB1*0201 haplotype, while the remaining patients do not carry this combination. Other loci might play a relevant role in the linkage of the HLA region with sarcoidosis, such as Tumor Necrosis Factor (TNF) genes. Numerous investigations of the TNF cluster have indicated that these locus/loci might contribute to disease susceptibility (10-13). A large number of biallelic polymorphisms have been identified within the TNF gene cluster, as well as 6 microsatellites (TNFa-TNFf). With the exception of TNFb microsatellite which is composed out of mononucleotide tandem repeats (G/A) n, TNF microsatellites are based on a dinucleotide repeat motif: TNFa (GT) n, TNFc (GA) n, TNFd (GA) n, TNFe (GA) n and TNFf (CA) n (13, 14). Data which support the inclusion of this region in the investigation of sarcoidosis was provided by various studies (15-18). At the same time, it is well documented that the microsatellites in the region between TNF gene cluster and DRB loci also play a role in the predisposition to some autoimmune diseases such as rheumatoid arthritis (RA), type 1 diabetes, Addison s disease as well as in psoriatic arthritis and scleroderma (19-22). An example of such a dinucleotide microsatellite locus is D6S273, which was a subject of various studies that reported about its role as an additional genetic marker (23-25). The D6S1014 microsatellite is another such example of association with RA, multiple sclerosis, schizophrenia, etc. (26-29). In the context of the possible role of TNF gene cluster in the etiology of sarcoidosis, and with respect to the possible association of microsatellites located between TNF and DRB1, the aim of the present study was to determine, by analyzing well-characterized subgroups (acute and chronic) of Croatian sarcoidosis patients, whether an association between sarcoidosis and HLA class I (A, B) and HLA class II (DRB1, DQB1) loci as well as with HLA microsatellites (TNFa, TNFb, TNFd, D6S273, and D6S1014) exists. Materials Patients and controls All the patients (N=133) included in this study were treated at the University Hospital for Lung Diseases Jordanovac, Zagreb. All patients were followed up for 2-18, average 4.5 years in the period Patients were recruited from Zagreb area and were all of Croatian origin. The University Hospital for Lung Diseases Jordanovac, Zagreb ethics committee approved the study. All patients signed their informed consent. The diagnosis of sarcoidosis was based on consistent clinical features, together with biopsy-proven noncaesating epitheloid cell granulomas according to the international guidelines (16). A summary of patient s characteristics is listed in Table 1. The control group (N=177) was comprised out of ethnically matched, healthy volunteer citizens of Zagreb, the capital city of Croatia. The entire investigated sample was previously typed for HLA specificities (7). Typing for HLA class I (A and B) was done by serology, while the DNA typing method as described in the published data was used for the determination of HLA class II (DRB1 and DQB1) by PCS-SSP (Polymerase Chain Reaction Sequence Specific Primers) (30). Methods Microsatellite typing DNA was extracted from EDTA anti-coagulated blood using a commercial kit (Nucleospin Blood, Macherey-Nagel, Duren, Germany).

3 20 Z. Grubić, T. Peroš-Golubičić, K. Štingl, R. Žunec Table 1. Patients characteristics n (%) Number of patients 133 Females 76 (57.1%) Males 57 (42.9%) Biopsy confirmation 100% Mean age at disease onset 42 yrs (23-65) Form of disease Acute (AcS) 93 (69.9%) With Erythema nodosum 43/93 With Erythema nodosum and Löfgren syndrome 19/93 Chronic (ChrS) 40 (30.1%) Pulmonary sarcoidosis 78 (48.6%) Extra-thoracic sarcoidosis 55 (41.4%) Chest radiograph Stage 0 66 (49.6% ) Acute/Chronic 34/32 Stage I 15 (11.3%) Acute/Chronic 4/11 Stage II 29 (21.8%) Acute/Chronic 9/20 Stage III 22 (16.5%) Acute/Chronic 2/20 Stage IV 1 (0.8%) Acute/Chronic 0/1 Lung function tests FVC < 80% 32 (24.1%9 FEV 1 < 80% 36 (27.1%) D LCO < 80% 65 (48.9%) FVC forced vital capacity; FEV 1 forced expiratory volume in one second; D LCO diffusing capacity for carbon monoxide Samples were amplified for 5 microsatellites (TNFa, TNFb, TNFd, D6S273, D6S1014). The amplification for each microsatellite was carried out in a separate reaction with 10 µl as a total reaction volume per sample. The forward primers were labeled with Cy5 fluorescent dye which enabled the laser detection of the amplified fragments. Primer sequences and conditions of amplification were as reported in a previously published study (31). Following the polymerase chain reaction (PCR) amplification, the PCR products were mixed with loading buffer and an internal standard, denatured and immediately placed on ice. Fragments were separated on a 6% polyacrylamide gel using an automated sequencer (ALFexpress; Pharmacia Biotech, Uppsala, Sweden). Allele identification was performed using fragment analysis software (AlleleLocator, Pharmacia Biotech). B cell lines from the 11 th International Histocompatibility Workshop (IHW) with known microsatellite alleles were included on each gel as additional internal standards: VAVY - TNFa2, TNFb3, TNFd1, D6S pb; TUBO-TNFa3/ TNFa13, TNFb1/b4, TNFc1/c2, TNFd4/d5; OMW TNFa7, TNFb1, TNFd2; IBW9-TNFa4, TNFb7, TNFc2, TNFd5; BTB D6S pb; TAB089 - D6S pb; BTB - D6S pb. Statistics Phenotype frequencies of the HLA specificities as well as the HLA microsatellite alleles were calculated for the patient s groups and control group. The chi-squared test was used to compare the expected value of genotype with its observed value in order to confirm whether it satisfies the Hardy-Weinberg law. The significance of associations between loci was estimated from 2X2 tables by chi-squared test with Yate s correction or by Fisher s exact test when the number of cases was less than five. The multiple comparisons were performed using Bonferroni correction for the number of comparisons made (P value was multiplied by 13 for A, 18 for B, 13 for DRB1, 14 for DQB1, 13 for TNFa, 5 for TNFb, 4 for TNFd, 7 for D6S273 and 7 for D6S1014). The odds ratio (OR) of disease risk for a given marker was also calculated from the 2X2 contingency table and a value of Pcorr<0.05 was considered as statistically significant. Results HLA class I analysis Among 14 HLA-A class I antigens observed in both subgroups (acute sarcoidosis AcS and chronic sarcoidosis ChrS) and twenty-one different antigens at HLA-B locus there were no statistically significant differences in comparison to control group (data not shown). HLA class II analysis The phenotype frequencies of HLA-DRB1 and DQB1 alleles are listed in Table 2. DRB1*03 specificity was significantly more present among the patients with the acute form of the disease in comparison to control subjects (31/ % vs. 19/ %; P=0.0005; Pcorr=0.0069) while in comparison to patients with chronic sarcoidosis the difference was not significant after correction (31/93-

4 The investigation of HLA microsatellites influence in predisposition to sarcoidosis among Croatians 21 Table 2. The distribution of DRB1 and DQB1 alleles among patients with sarcoidosis and controls HLA- Acute sarcoidosis Chronic sarcoidosis Controls n % PF n % PF n % PF DRB1* N=93 N=40 N= DQB1* N=93 N=40 N= , Legend: PF-phenotype frequency; 1 Patients with acute sarcoidosis vs. controls P=0.0414, Pcorr>0.05; 2 Patients with acute sarcoidosis vs. controls P=0.0005; Pcorr=0.0069; 2 Patients with acute sarcoidosis vs. patients with chronic sarcoidosis P=0.0354, Pcorr>0.05; 3 Patients with chronic sarcoidosis vs. controls P=0.0219; Pcorr>0.05; 4 Patients with chronic sarcoidosis vs. controls P=0.0156, Pcorr>0.05; 4 Patients with chronic sarcoidosis vs. patients with acute sarcoidosis P=0.0181, Pcorr> % vs. 6/ %; P=0.0354; Pcorr>0.05). The increase of the DRB1*11 frequency among patients with chronic sarcoidosis was observed in comparison to controls (21/ % vs. 44/ %; P=0.0219; Pcorr>0.05; OR=2.44) but not in comparison to patients with acute sarcoidosis (21/ % vs. 31/ %; P>0.05). The situation was the same with the frequency of DRB1*15 specificity, which was also more present among ChrS patients in comparison to the control subjects (13/ % vs. 23/ %, P=0.0414), but not in comparison to the patients with AcS. At the same time, the lower frequency of DRB1*13 specificity among ChrS group revealed a difference in comparison to AcS (P=0.0181; Pcorr>0.05) as well as to controls (P=0.0156; Pcorr>0.05; OR=0.24). None of the other DRB1 specificities has reached a statistically significant difference in the frequency when the tested groups were compared. At DQB1 locus, only DQB1*0201 allele showed a significantly higher frequency among AcS patients in comparison to controls (21/ % vs. 16/ %; P=0.0009; Pcorr=0.0117) but not to ChrS patients (21/ % vs. 6/ %; P=0.0354; Pcorr>0.05). Analysis of the presence of the HLA- DRB1*03/DQB1*0201 haplotype among patients with sarcoidosis showed that this haplotype is more present in cases of the acute form of the disease than in subgroup of patients with the chronic form of sarcoidosis (P=0.0354; Pcorr>0.05; OR=2.83). This difference was significant in comparison to controls (P=0.0001; Pcorr=0.0168; OR=3.22).

5 22 Z. Grubić, T. Peroš-Golubičić, K. Štingl, R. Žunec HLA microsatellite analysis The distribution of TNFa microsatellite alleles was also determined in the same groups of patients and controls. No significant differences were found between the two subgroups of patients. A tendency to an increased frequency was found for TNFa-4 allele among patients with acute sarcoidosis (19/ %) in contrast to the group of patients with the chronic form of the disease (3/40-7.5%), whereas TNFa-11 allele was decreased (13/ % vs. 9/ %; P>0.05). The comparison of the AcS patients group versus ChrS patients group did not reveal significant differences for any of the TNFb alleles. Compared with the controls, a slightly increased frequency of TNFb-3 allele in the group of patients with AcS was found (34/ % vs. 37/ %; P=0.0350; Pcorr>0.05). At the TNFd locus the comparison of patient and control groups showed a statistically significant increase of the TNFd-2 allele frequency among patients with AcS (37/ % vs. 20/ %; P< ; Pcorr = ), while the increase of this allele s frequency was not statistically significant among patients with ChrS (12/ % vs. 20/ %; P =0.0091; Pcorr >0.05). The distribution of all other alleles at TNFd locus did not differ from the distribution found among controls. The distribution of D6S273 allele frequencies showed that D6S273-7 allele is significantly more present among patients with AcS in comparison to control subjects (22/ % vs. 17/ %; P=0.0003; Pcorr=0.0021), while the difference found for patients with ChrS was not statistically significant (7/ % vs. 17/ %; P>0.05). The low frequency of D6S273-3 allele among patients with ChrS did not reach a significant P value in comparison of the frequency of this allele among controls (3/33-9.1% vs. 54/ %; P=0.0100; Pcorr>0.05), as well as in comparison to patients with AcS. The difference in the frequency of D6S273-3 allele among patients with AcS and controls was also observed but without statistical significance after correction (17/ % vs. 54/ %; P=0.0424; Pcorr>0.05). The last analyzed microsatellite locus was D6S1014. Four out of eight alleles known so far (D6S1014-1, D6S1014-4, D6S1014-7, and D6S1014-8) demonstrated a difference in frequencies between patients groups and controls. Only the frequency of D6S allele varied between two subgroups of patients (AcS: 6/93-6.5% vs. ChrS: 6/ %) but without a significant P value. At the same time, this allele was observed with a significantly higher frequency among ChrS patients in comparison to the controls (6/ % vs. 3/ %; P=0.0037; Pcorr=0.0296). The D6S allele was less present among AcS patients in comparison to controls (45/ % vs. 93/ %; P=0.0219; Pcorr>0.05), but also among patients with ChrS in comparison to healthy individuals (17/ % vs. 93/ %; P=0.0223; Pcorr>0.05). Contrary to this allele, D6S allele was increased among patients with AcS (23/ % vs. 12/ %; P=0.0004; Pcorr=0.0028) and patients with ChrS (9/ % vs. 12/ %; P=0.0151; Pcorr>0.05). The presence of D6S allele was slightly lower among both subgroups of patients in comparison to the control group. Extended HLA haplotype analysis The final aim of the study was to investigate the extended haplotypes among the tested groups. This analysis showed that there were no significant differences between patient groups as well as in comparison of patients with the control group. With respect to the HLA-DRB1 specificities that have shown significant difference between patient s groups or one patient group and controls, the following haplotypic associations were observed: HLA- DRB1*03/DQB1*0201/TNFa3/TNFb3/TNFd2/ D6S273-7/D6S (AcS: 25/ % vs. ChrS: 4/ %; P=0.0387; Pcorr>0.05 and AcS: 25/ % vs. controls: 15/ %; P=0.0019; Pcorr>0.05), HLA-DRB1*11/DQB1*0301/TNFa 10/TNFb4/TNFd4/D6S273-4/D6S (ChrS: 12/ % vs. AcS: 12/ %; P=0.0353; Pcorr>0.05), and HLA-DRB1*13/DQB1*06/ TNFa10/TNFb1/TNFd5/D6S273-4/D6S (AcS: 14/ % vs. ChrS: 1/40-2.5%; P=0.0384; Pcorr>0.05). There were no statistically significant differences between patient s groups or one patient group and controls when we included HLA class I specificities in extended haplotypic association analysis. This was the case for all previously mentioned hap-

6 The investigation of HLA microsatellites influence in predisposition to sarcoidosis among Croatians 23 Table 3. The distribution of TNFa, TNFb, TNFd, D6S273 and D6S1014 alleles among patients with sarcoidosis and controls Microsatellite Acute sarcoidosis Chronic sarcoidosis Controls n % PF n % PF n % PF TNFa (bp) N=93 N=40 N=177 1 (97) (99) (101) (103) (105) (107) (109) (111) (113) (115) (117) (119) (121) (123) TNFb (bp) N=91 N=39 N=155 1 (124) (125) (126) (127) (128) (129) (130) TNFd (bp) N=93 N=40 N=168 1 (124) (126) (128) (130) (132) (134) (136) D6S273 (bp) N=88 N=33 N=168 1 (126) (128) (130) (132) (134) (136) (138) D6S1014 (bp) N=93 N=40 N=145 1 (125) (128) (131) (134) (137) (140) (143) (146) Legend: PF-phenotype frequency; 1 Patients with acute sarcoidosis vs. controls P= , Pcorr>0.05; 2 Patients with acute sarcoidosis vs. controls P= ; Pcorr= ; 2 Patients with chronic sarcoidosis vs. controls P=0.0091, Pcorr>0.05; 3 Patients with acute sarcoidosis vs. controls P=0.0424, Pcorr>0.05; 3 Patients with chronic sarcoidosis vs. controls P=0.0100, Pcorr>0.05; 4 Patients with acute sarcoidosis vs. controls P=0.0003; Pcorr>0.0021; 5 Patients with acute sarcoidosis vs. controls P=0.0219, Pcorr>0.05; 5 Patients with chronic sarcoidosis vs. controls P=0.0223, Pcorr>0.05; 6 Patients with acute sarcoidosis vs. controls P=0.0004, Pcorr=0.0028; 6 Patients with chronic sarcoidosis vs. controls P=0.0151, Pcorr>0.05; 7 Patients with chronic sarcoidosis vs. controls P=0.0037, Pcorr>0.0296;

7 24 Z. Grubić, T. Peroš-Golubičić, K. Štingl, R. Žunec lotypic associations which included only HLA class II region. Even the increase in frequency of HLA- A1/B8/DRB1*03/DQB1*0201/TNFa3/TNFb3/ TNFd2/D6S273-7/D6S among patients with acute form of disease (17/ %) did not reach statistically significant P value in comparison to patients with chronic form of sarcoidosis (4/40-10.%) nor controls (10/ %). The difference was even less pronounced for haplotypic associations: HLA- 2/B18/DRB1*11/DQB1*0301/TNFa10/TNFb4/ TNFd4/D6S273-4/D6S (AcS: 5/93-5.4%; ChrS: 3/40-7.5%; controls: 12/ %) HLA- A2/B44/DRB1*13/DQB1*06/TNFa10/TNFb1/ TNFd5/D6S273-4/D6S (AcS: 4/93-4.3%; ChrS: 0/40-0%; controls: 3/ %). Discussion The research on the association between sarcoidosis and HLA region has been carried out for a long period of time, but many questions are still open. A consensus about which HLA locus is directly involved in the pathogenesis of sarcoidosis has not been established to date. In our previous case control study in the Croatian population we reported about the association between HLA- B8/DRB1*0301/DQB1*0201 haplotype with acute onset of disease, radiological stage I, erythema nodosum, Löfgren syndrome, no-medical therapy, and pulmonary sarcoidosis. The present study is an extension of the previous study with the aim of determining which genetic markers within the HLA region are associated with acute or chronic form of disease. Consistent with previous report, we have demonstrated the association of HLA-DRB1*03/DQB1*0201 haplotype with the acute form of the disease. This confirms the findings from United Kingdom, Poland, Czech Republic, Italy, and Germany which reported about the HLA- DRB1*0301 association with a good prognosis and the acute form of the disease (32-35). At the same time, we did not find a correlation between HLA genes or haplotypes with the chronic form of sarcoidosis. Namely, the higher frequency of HLA-DRB1*11 specificity in this subgroup of sarcoidosis patients was significant only before correction. This is similar to the findings from Poland, India and USA who have also reported about an association between DRB1*11 specificity and the chronic form of sarcoidosis (36, 37). Simultaneously, various groups of authors reported about a higher frequency of HLA-DRB1*1501 and DRB1*14 among patients with chronic sarcoidosis (38-40). HLA- DRB1*15 specificity was also increased in our sample of patients with chronic form of sarcoidosis but only before correction of the P value, while the higher frequency of DRB1*14 specificity did not reach a statistically significant P value even before correction. A lower frequency of DRB1*13 specificity among patients with the chronic sarcoidosis but only before correction, was also observed in the Croatian population. All these differences could be explained by the fact that the association of HLA- DRB1 specificities with sarcoidosis activation and disease outcome is ethnically dependent; on the other hand, this could also be explained by a smaller number of patients in our study which affected the statistical power of analysis. The results obtained after analysis of HLA- DQB1 among patients and controls are in good concordance with previous reports with respect to DQB1*0201 which is in association with DRB1*0301. On the other hand, no such significant association was found for DQB1*0602 and persistent disease as was reported by other authors (41). This correlates well with our finding for DRB1*15 which is in linkage disequilibrium with this DQB1 allele and can also be explained by a lower statistical power of analysis due to relatively small group of patients. This study is not the first attempt to discover other genetic markers which play a role in the immunopathogenesis of sarcoidosis. A few studies have reported about the association between TNF polymorphisms and sarcoidosis (11, 14-16). More precisely, a higher frequency of the TNF-307A allele was found among patients with Löfgren s syndrome, while a study from the Czech Republic reported about an association between two other single nucleotide polymorphisms (SNPs) and Löfgren s syndrome (TNF-308*G/A and LTA*252*A/G). Additional support for the investigation of the polymorphisms within TNF gene cluster was given by a Japanese study in which the authors reported about the association between TNFa2 allele and cardiac sarcoidosis (17, 18). The clinical course of sarcoidosis has already been associated with patient s individ-

8 The investigation of HLA microsatellites influence in predisposition to sarcoidosis among Croatians 25 ual capacity of spontaneous TNF-α production by alveolar macrophages (42). The protein level of this cytokine has been shown to correlate with the polymorphisms located within the regulatory region of its gene, especially with the TNF-308 polymorphism (43). According to our knowledge, this is the first time that the polymorphisms of TNF microsatellites were analyzed among patients with sarcoidosis. Our data did not reveal any association between TNFa and TNFb microsatellite alleles and sarcoidosis. On the other hand, the only statistically significant difference concerning the TNFd locus, even after the correction of P value, was observed for TNFd-2 allele. The increase of the TNFd-2 allele frequency was found for both subgroups of patients. While among patients with AcS this result could be interpreted with respect to the fact that this allele is also a part of 8.1 AH (ancestral haplotype), the fact that HLA-B*08 specificity, as well as DRB1*0301 allele frequencies were not increased among patients with ChrS suggests that this TNFd allele might be considered as an independant risk factor for chronic sarcoidosis. Regarding the increased frequency of D6S273-7 allele among patients with AcS in comparison to the controls, after the stratification analysis was performed for this allele, we can conclude that this increase is a result of linkage disequilibrium of this allele with DRB1*03 specificity. At the same time, on the base of our results we can suggest that D6S273-3 allele is protective for the chronic form of sarcoidosis (OR=0.21). Considering that the HLA-DRB1 locus is in linkage disequilibrium with D6S1014 locus, we have also performed a stratification analysis taking into account polymorphisms at these loci. Results demonstrated that the higher frequency of D6S allele and odds ratio (OR=8.35) are not a result of LD with DRB1*11 or DRB1*15 which have also shown a positive correlation with ChS (OR=2.44 and OR=2.47, respectively). Our data suggest that D6S allele independently contributes to the susceptibility to chronic sarcoidosis. On the base of our results we can hypothesize that gene(s) associated with sarcoidosis is located between D6S273 and D6S1014 microsatellites in the HLA class III region (G6D gene) (43). This gene is a member of the leukocyte antigen 6 superfamily, which is important in leukocyte maturation and for that reason it might be very helpful to screen sarcoidosis patients for their polymorphisms. Haplotypic association analysis which included markers located in HLA class II region demonstrated positive association HLA-DRB1*03/DQB1* / T N F a 3 / T N F b 3 / T N F d 2 / D 6 S /D6S and HLA-DRB1*13/DQB1*06/ TNFa10/TNFb1/TNFd5/D6S273-4/D6S combinations with acute sarcoidosis, while for patients with chronic form of sarcoidosis a positive association was found for the HLA-DRB1*11/ DQB1*0301/TNFa10/TNFb4/TNFd4/D6S273-4/D6S combination. No statistically significant association was found for any of the haplotypic combinations which included the HLA class I region which supports the hypothesis that genetic markers in HLA class II region are more involved in predisposition to this disease than those located in the HLA class I region. In conclusion, the present study identified non- HLA markers associated with sarcoidosis giving a new point of view on the genetic background for this disease and suggesting the involvement of the region between HLA-DQB1 and D6S273 loci in its pathophysiology. Acknowledgment This study was supported by the Grant ( ) of the Ministry of Science, Sport and Education of the Republic of Croatia. References 1. Baughman R, Lower E, du Bois RM. Sarcoidosis. Lancet 2003; 361: Ziegenhagen MW, Müller-Quernheim J. The cytokine network in sarcoidosis and its clinical relevance. J Intern Med. 2003; 253: Martinetti M, Luisetti M, Cuccia M. HLA and sarcoidosis: new pathogenetic insights. Sarcoidosis Vasc Diffuse Lung Dis 2002; 19: Sharma SK, Balamurugan A, Pandey RM. et al. Human Leukocyte Antigen-DR alleles influence the clinical course of pulmonary sarcoidosis in Asian Indians. Am J Respir Cell Mol Biol 2003; 29: Rossman MD, Thompson B, Frederick M, et al. HLA-DRB1*1101: A sigificant risk factor for sarcoidosis in Blacks and Whites. Am J Hum Genet 2003; 73: Schürmann M, Lympany PA, Reichel P, et al. Familial sarcoidosis is linked to the Major Histocompatibility Complex Region. Am J Respir Crit Care Med 2000; 162: Grubić Z, Žunec R, Peroš-Golubičić T, et al. HLA class I and class II frequencies in patients with sarcoidosis from Croatia: role of HLA-B8,

9 26 Z. Grubić, T. Peroš-Golubičić, K. Štingl, R. Žunec -DRB1*0301, and DQB1*0201 haplotype in clinical variations of disease. Tissue Antigens 2007; 70: Swider C, Schnittger L, Bogunia-Kubik K, et al. TNF-alpha and HLA-DR genotyping as potential prognostic markers in pulmonary sarcoidosis. Eur Cytokine Netw 1999; 10: Seitzer U, Gerdes J, Müller-Quernheim. Evidence of disease phenotype associated haplotypes (DR.TNF) in sarcoidosis. Sarcoidosis Vasc Diffuse Lung Dis 2001; 18: Somoskovi A, Zissel G, Scitzer U, Gerdes J, Schlaak M, Müller- Quernheim J. Polymorphisms at position 308 in the first intron of the TNF-beta genes and spontaneous and lipopolysaccharide-induced TNF-alpha release in sarcoidosis. Cytokine 1999; 11: Pandey JP, Frederick M, and ACCESS research group. TNF-α, ILβ, and Immunoglobulin (GM and KM) gene polymorphisms in sarcoidosis. Hum Immunology 2002; 63: Udalova IA, Nedosasova SA, Webb GC, Chaplin DD, Teretskaya RL. Higly informative typing the human TNF locus using six adjacent polymorphic markers. Genomics 1993; 16: Tsukamoto K, Ohta N, Shirai Y, Emi M. A highly polymorphic CA repeat marker at the human tumor necrosis factor alpha (TNFα) locus. J Hum Genet 1998; 43: Mrazek F, Holla LI, Hutyrova B, et al. Association of tumour necrosis factor-α, lymphotoxin-α and DRB1 gene polymorphisms with Löfgren s syndrome in Czech patients with sarcoidosis. Tissue Antigens 2004; 65: Labunski S, Posern G, Ludwig S, Kundt G, Bröcker EB, Kunz M. Tumour necrosis factor-alpha promoter polymorphism in erythema nodosum. Acta Derm Venereol 2001; 81: Grutters JC, Sato H, Pantelidis P, et al. Increased frequency of the uncommon tumor necrosis factor -857T allele in British and Dutch patients with sarcoidosis. Am J Respir Crit Care Med 2002; 165: Takashige N, Naruse TK, Matsumori A, et al. Genetic polymorphism of tumor necrosis factor-β but not tumor necrosis factor loci (TNFA and TNFB) in cardiac sarcoidosis. Tissue Antigens 1999; 54: Haajer AH, Hutchunson IV. Influence of TNFα gene polymorphisms on TNFα production and disease. Hum Immunology 2001; 62: Mattey DL, Hassell AB, Dawes PT, Ollier WER, Haajer A. Interaction between tumor necrosis factor microsatellite polymorphisms and the HLA-DRB1 shared epitope in rheumatoid arthritis. Arthritis & Rheumatism 1999; 42: Törn C, Hillman M, Sanjeevi CB, Landin-Olsson M. Polymorphisms of TNF microsatellite marker a and HLA-DR-DQ in diabetes mellitus-a study in 609 Swedish subjects. Hum Immunol. 2006; 67: Frank KH, Füssel M, Conrad K, et al. Different distribution of HLA class II and tumor necrosis factor alleles (TNF-308.2, TNFa2 microsatellite) in anti-topoisomerase I responders among scleroderma patients with and without exposure to quartz/metal dust. Arthritis Rheum 1998; 41: Alenius GM, Friberg C, Nilsson S, Wahlström J, Dahlqvist SR, Samuelsson L. Analysis of 6 genetic loci for disease susceptibility in psoriatic arthritis. J Rheumatol 2004; 31: Valdes AM, Wapelhorst B, Concannon P, Erlich HA, Thomson G, Noble JA. Extended DR3-D6S273-HLA-B haplotypes are associated with increased susceptibility to type 1 diabetes in US Caucasians. Tissue Antigens 2005; 65: Pascual M, Matarán L, Jones G, et al. HLA haplotypes and susceptibility to rheumatoid arthritis. More than class II genes. Scand J Rheumatol 2002; 31: Harjaek M, Margeti T, Kerhin-Brkljaci V, Martinez N, Grubi Z. HLA-B*27/HLA-B*07 in combination with D6S allele is associated with increased susceptibility to juvenile spondyloarthropathies. Clin Exp Rheumatol 2008; 26: van de Beek WJ, Roep BO, van der Slik AR, Giphart MJ, van Hilten BJ. Susceptibility loci for complex regional pain syndrome. Pain 2003; 103: International Psoriasis Genetics Consortium.The International Psoriasis Genetics Study: assessing linkage to 14 candidate susceptibility loci in a cohort of 942 affected sib pairs. Am J Hum Genet. 2003; 73: de Jong BA, Huizinga TW, Zanelli E, et al. Evidence for additional genetic risk indicators of relapse-onset MS within the HLA region. Neurology 2002; 59: Gourraud PA, Mano S, Barnetche T, Carrington M, Inoko H, Cambon-Thomsen A. Integration of microsatellites characteristics in the MHC region: a literature and sequence based analysis. Tissue Antigens 2004; 64: Grubić Z, Žunec R, Čečuk-Jeličić E, Kerhin-Brkljačić V, Kaštelan A. Polymorphism of HLA-A, B, DRB1, DQA1 and DQB1 haplotypes in the Croatian population. Eur J Immunogenet 2000; 27: Grubić Z, Štingl K, Žunec R, Car H, Čečuk-Jeličić E, Kerhin-Brkljačić V. Linkage disequilibrium between human leukocyte antigen-b and closely related microsatellites in the Croatian population. Tissue Antigens 2006; 89: Sato H, Grutters JC, Pantelidis P, et al. HLA-DQB1*0201: a marker for good prognosis in British and Dutch patients with sarcoidosis. Am J Respir Cell Mol Biol 2002; 27: Bogunia-Kubik K, Tomeczko J, Suchnicki K, Lange A. HLA- DRB1*03, DRB1*11or DRB1*12 and their respective DRB3 specificities in clinical variants of sarcoidosis. Tissue Antigens 2001; 57: Spagnolo P, du Bois RM. Genetics of sarcoidosis. Clinicis in Dermatology 2007 ; 25 : Grunewald J, Brynedal B, Darlington P, et al. Different HLA-DRB1 allele distributions in distinct clinical subgroups of sarcoidosis patients. Respiratory Research, 2010; 11: Rosman MD, Thompson B, Frederick M, et al. HLA-DRB1*1101: A significant risk factor for sarcoidosis in Blacks and Whites. Am J Hum Genet 2003; 73: Dubaniewicz A, Dubaniewicz-Wybieralska M, Moszkowska G, Sternau A, Dubaniewicz A. Comparative analysis of DR and DQ alleles occurrence in sarcoidosis and tuberculosis in the same ethnic group: preliminary study. Sarcoidosis Vasc Diffuse Lung Dis 2006; 23: Sharma SK, Balamurugan A, Pandey RM, Saha PK, Mehra NK. Human Leukocyte Antigen-DR alleles influence the clinical course of pulmonary sarcoidosis in Asian Indians. Am J Respir Cell Moll Biol 2003; 29: Rutherford RM, Brutsche MH, Kearns M, Bourke M, Stebens F, Gilmartin JJ. HLA-DR2 predicts susceptibility and disease chronicity in Irish sarcoidosis patients. Sarcoidosis Vasc Diffuse Lung Dis 2004; 21: Voorter CEM, Drent M van den Berg-Loonen EM. Severe sarcoidosis is strongly associated with the haplotype HLA- DQB1*0602-DRB1* Hum Immun 2005; 66: Grunewald J, Eklund A, Olerup O. Human leukocyte antigen class I alleles and the disease course in sarcoidosis patients. Am J Respir Crit Care Med 2004; 169: Hino T, Nakamura H, Shibata Y, Abe S, Kato A, Tomoike H. Elevated of type II soluble tumor necrosis factor receptors in the bronchoalveolar lavage fluide of patients with sarcoidosis. Lung 1997; 175: Stuber P, Petersen M, Bokelmann F, Schade U. A genomic polymorphism within the tumor necrosis factor locus influences plasma tumor necrosis factor-α concentrations and outcome of patients with severe sepsis. Cri Care Med 1996; 24: